Selecting Content Using Entity Properties

1. A method comprising: receiving, by one or more processors, a search query that refers to a first entity; generating, by the one or more processors using the search query and an entity graph that includes one or more relationship connections between entities in the entity graph, a search query graph that comprises a plurality of nodes, wherein the search query graph includes: nodes respectively representing each of: (a) a first entity included in the search query; and (b) a second entity that is beyond what is included in the received query, but is connected to the first entity in the entity graph by a relationship connection that specifies how the first entity and the second entity are related; and wherein a topology of the search query graph is based on the entity graph; identifying, from a database, a particular content selection criteria graph having the topology that matches the topology of the search query graph generated using the entity graph, wherein the identifying comprises performing a node-by-node comparison of the search query graph with each of one or more candidate content selection criteria graphs that indicate selection criteria specified by one or more content providers different from a provider of the entity graph and provided by the one or more content providers different from the provider of the entity graph; identifying a set of content items that are distributed when the particular content selection criteria graph is matched, wherein the content items are provided by a particular content provider of the one or more content providers that provided the particular content selection criteria graph; and providing, by the one or more processors and for display on a computing device that submitted the search query, one or more from among the identified set of content items based on the identified match between the particular content selection criteria graph and the search query graph.

2. The method of claim 1 , wherein identifying the content selection criteria graph further comprises determining whether a topology of the search query graph matches a topology of the one or more candidate content selection criteria graphs.

3. The method of claim 1 , wherein identifying the content selection criteria graph further comprises comparing one or more of structures, relationships, or links of the search query graph and the one or more candidate content selection criteria graphs.

4. The method of claim 1 , wherein identifying content items associated with the particular content selection criteria graph comprises identifying one of a content campaign or content group associated with the particular content selection criteria graph.

5. The method of claim 1 , wherein identifying the content selection criteria graph further comprises determining an entity confidence score satisfies an entity threshold.

6. The method of claim 5 , wherein the entity confidence score represents a semantic relevancy of a particular entity to the search query.

7. The method of claim 5 , wherein the confidence score is based on browsing history data.

8. The method of claim 1 , further comprising: identifying, from the search query, the first entity from a data structure that stores entity information; and identifying a confidence score for the first entity, wherein the confidence score represents a semantic relevancy of the first entity to the search query; wherein identifying the content selection criteria graph further comprises determining that the confidence score satisfies an entity threshold.

9. The method of claim 1 , further comprising conducting an auction using bids submitted by content providers of the identified content items.

10. A system comprising: one or more computer processors; and one or more non-transitory computer readable devices that include instructions that, when executed by the one or more computer processors, cause the one or more computer processors to perform operations, the operations comprising: receiving a search query that refers to a first entity; generating, using the search query and an entity graph that includes one or more relationship connections between entities in the entity graph, a search query graph that comprises a plurality of nodes, wherein the search query graph includes: nodes respectively representing each of: (a) a first entity included in the search query; and (b) a second entity that is beyond what is included in the received query, but is connected to the first entity in the entity graph by a relationship connection that specifies how the first entity and the second entity are related; and wherein a topology of the search query graph is based on the entity graph; identifying, from a database, a particular content selection criteria graph having the topology that matches the topology of the search query graph generated using the entity graph, wherein the identifying comprises performing a node-by-node comparison of the search query graph with each of one or more candidate content selection criteria graphs that indicate selection criteria specified by one or more content providers different from a provider of the entity graph and provided by the one or more content providers different from the provider of the entity graph; identifying a set of content items that are distributed when the particular content selection criteria graph is matched, wherein the content items are provided by a particular content provider of the one or more content providers that provided the particular content selection criteria graph; and providing, for display on a computing device that submitted the search query, one or more from among the identified set of content items based on the identified match between the particular content selection criteria graph and the search query graph.

11. The system of claim 10 , wherein identifying the content selection criteria graph further comprises determining whether a topology of the search query graph matches a topology of the one or more candidate content selection criteria graphs.

12. The system of claim 10 , wherein identifying the content selection criteria graph further comprises comparing one or more of structures, relationships, or links of the search query graph and the one or more candidate content selection criteria graphs.

13. The system of claim 10 wherein identifying content items associated with the particular content selection criteria graph comprises identifying one of a content campaign or content group associated with the particular content selection criteria graph.

14. The system of claim 10 , wherein identifying the content selection criteria graph further comprises determining an entity confidence score satisfies an entity threshold.

15. The system of claim 14 , wherein the entity confidence score represents a semantic relevancy of a particular entity to the search query.

16. The system of claim 14 , wherein the confidence score is based on browsing history data.

17. The system of claim 10 , the operations further comprising: identifying, from the search query, the first entity from a data structure that stores entity information; and identifying a confidence score for the first entity, wherein the confidence score represents a semantic relevancy of the first entity to the search query; wherein identifying the content selection criteria graph further comprises determining that the confidence score satisfies an entity threshold.

18. The system of claim 10 , the operations further comprising conducting an auction using bids submitted by content providers of the identified content items.

19. A non-transitory computer-readable medium comprising processor executable instructions to select content via a computer network, the instructions further comprising instructions to: receive a search query that refers to a first entity; generate, using the search query and an entity graph that includes one or more relationship connections between entities in the entity graph, a search query graph that comprises a plurality of nodes, wherein the search query graph includes: nodes respectively representing each of: (a) a first entity included in the search query; and (b) a second entity that is beyond what is included in the received query, but is connected to the first entity in the entity graph by a relationship connection that specifies how the first entity and the second entity are related; and wherein a topology of the search query graph is based on the entity graph; identify, from a database, a particular content selection criteria graph having the topology that matches the topology of the search query graph generated using the entity graph, wherein the identifying comprises performing a node-by-node comparison of the search query graph with each of one or more candidate content selection criteria graphs that indicate selection criteria specified by one or more content providers different from a provider of the entity graph and provided by the one or more content providers different from the provider of the entity graph; identify a set of content items that are distributed when the particular content selection criteria graph is matched, wherein the content items are provided by a particular content provider of the one or more content providers that provided the particular content selection criteria graph; and provide, for display on a computing device that submitted the search query, one or more from among the identified set of content items based on the identified match between the particular content selection criteria graph and the search query graph.

20. The non-transitory computer-readable medium of claim 19 , the instructions further comprising instructions to: identify, from the search query, the first entity from a data structure that stores entity information; and identify a confidence score for the first entity, wherein the confidence score represents a semantic relevancy of the first entity to the search query; wherein the instructions to identify the content selection criteria graph further comprise instructions to determine that the confidence score satisfies an entity threshold.